Lane-Level Traffic Prediction Using Local Pattern Caching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for autonomous vehicles and advanced driver-assistance systems require real-time lane-level traffic information, which is hindered by limited network connectivity and bandwidth, leading to inadequate navigation performance when real-time data is unavailable.
Innovation Solution
The method involves generating and incorporating lane direction patterns, divergent parameter patterns, and lane event patterns into digital maps, allowing vehicles to predict traffic conditions and travel times using probe data from vehicle apparatuses, even without continuous network connectivity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If real-time lane-level traffic information is obtained through network connection, then navigation accuracy is improved, but network bandwidth consumption increases and system reliability deteriorates when network is unavailable
Solution Approach 1:
The system pre-generates and stores traffic pattern data during periods when network connectivity is available. This preliminary action creates a local cache of traffic information that can be accessed offline, ensuring navigation accuracy is maintained even when network connection is unavailable, thus resolving the reliability issue without sacrificing measurement precision
Solution Approach 2:
The patent introduces an intermediary layer between real-time network data and navigation functions. This intermediary processes and stores traffic patterns locally, allowing the system to function reliably offline while maintaining accuracy. The intermediary acts as a buffer that decouples the dependency on continuous network connectivity
2Loss of information
If lane-level traffic information is continuously updated in real-time, then traffic information accuracy is improved, but network bandwidth consumption increases
Solution Approach 1:
Instead of continuously updating all traffic information in real-time, the system applies partial updates only when necessary. Traffic patterns are updated based on significant changes or at optimized intervals, reducing network bandwidth consumption while maintaining sufficient traffic information accuracy for navigation purposes
Solution Approach 2:
The system extracts and stores essential traffic pattern characteristics locally, separating the critical navigation-related information from the complete real-time traffic data stream. This extraction allows the system to maintain accurate traffic information for navigation while minimizing network bandwidth consumption by not continuously transmitting all available traffic data
Data Source
AI summary
A plurality of instances of probe data are received. Each instance of probe data corresponds to travel of a vehicle apparatus along a first segment, comprises an indication of at least one parameter characterizing the travel of the vehicle apparatus along the first segment. The at least one parameter is extracted from the instances of probe data to generate a distribution of parameters. One or more clusters of instances of probe data are identified based on the distribution of parameters. Responsive to identifying two or more clusters, a representative at least one parameter is determined for each cluster, and an element of a data structure is modified based thereon for each cluster. Responsive to identifying only one cluster of instances of probe data, the data structure is not modified based on the cluster. A navigation application is configured to use the data structure to perform a navigation function.